t f p g+ YouTube icon

Miracles are Impossible, Part 1

Bookmark and Share

June 24, 2013 Tags: Divine Action & Purpose
Miracles are Impossible, Part 1

Today's entry was written by Jeffrey Burton Russell. Please note the views expressed here are those of the author, not necessarily of BioLogos. You can read more about what we believe here.

Note: Today’s post comes to BioLogos as an excerpt from Jeffrey Burton Russell’s book, Exposing Myths About Christianity: A Guide to Answering 145 Viral Lies and Legends. Russell is Professor Emeritus of History at the University of California, Santa Barbara, and in this text, he critically examines numerous “viral lies and legends” about Christianity that have spread through popular culture in recent years. The idea that miracles are preposterous lies is a favorite attack against faith by the likes of the New Atheists, among others. Russell effectively dismantles these attacks by explaining various ways in which, “The statement that nothing can exist that can’t be explained by science is not a scientific statement.”

122. “Miracles are explained away by science.”

The essence of Christianity is based on belief in two overwhelming miracles: the creation of the cosmos and the resurrection of Jesus Christ. Neither of these can be explained away by science. A miracle is a supernatural event, which in its very nature is inaccessible to science, and any scientist honestly open to inquiry into natural phenomena—which is her vocation—will admit as much. The statement that nothing can exist that can’t be explained by science is not a scientific statement. “What is the philosophical argument or scientific experiment that conclusively assigns to miracles this supreme degree of unlikelihood?”[1] There isn’t one. Some atheists claim that miracles would make science impossible on the grounds that if there is a God who can intervene in natural events there can be no regularity in natural events.[2] But the rare suspension of natural regularities does not mean that natural regularities cannot exist. Some say that God should submit himself to the scientific tests of people in the twenty-first century. Aside from the obvious fact that God cannot be scientifically tested, this idea of twenty-first century intellectual elitists is chronological ethnocentricity, the arrogant assumption that “we” know better about everything than other people at other times. The idea of the regularities of God’s ordered universe is compatible with science; and the Christian idea of such regularities is one of the most important sources of science. People who believe in miracles usually don’t think of them as God’s “breaking” the laws of nature but as God’s being constantly attentive to his cosmos.

Antitheists argue that a universe made by God would be different from one made by only natural occurrences. To them, the universe appears to have no signs of being made by a planner. But there are such signs, in the order of the universe, in the direction of time, and the existence of miracles. The antitheists simply exclude the evidence without examining it.[3]

Statistics show that most Americans believe in miracles. Elitist atheists reply that most Americans are dolts. Of people 45 years of age and over, 80% believe in miracles, 41% believe that they occur every day, and 37% claim to have witnessed at least one. 85% of women believe, and 73% of men (males being more inclined to the materially practical). 86% of people with high school diplomas believe, and 71% with college degrees. As with other religious indicators, the big division is financial: 86% of people making less than $25,000 a year believe, as contrasted with 78% of those making $75,000 or more.[4] The more comfortably off one is, the less one is likely to believe in miracles—or anything supernatural. But it may be that folks working hard to support their families are more connected with actual reality than people patting one another on the back in expensive restaurants for being so much better than the folks serving them.

The atheist Michael Shermer tried to reconcile science and miracles through “the Law of Large Numbers,” which says that “an event with a low probability of occurrence in a small number of trials has a high probability of occurrence in a large number of trials…. Events with a-million-to-one odds happen 295 times a day in America.”[5] This view mistakes marvels or wonders for miracles. Strange marvels have occurred, such as rains of frogs, but these eventually have a scientific explanation. Small wonders are scientifically possible, but a gigantic wonder such as the resurrection of a really dead person is outside the boundaries of science. Shermer’s view relies upon statistics about wonders, but miracles are immune to statistics. There are no odds on miracles. Scientifically the odds against a resurrection are ∞:1, but religiously the chance of a miraculous resurrection is 0:0 or x:x. There is no way of predicting any odds of its occurring. It either occurs—or not. Therefore when one prays for a miracle one is not playing against the odds. Whether a miracle happens or not is entirely out of the realm of prediction. Christians, by the way, are usually cautious in reporting a miracle because reported miracles are usually mere marvels, hallucinations, or frauds.

Christian theology doesn’t say that “God can do anything,” but “God can do anything that isn’t self contradictory.” An old question is whether God can create a stone heavier than he can lift. The answer is of course not, and this answer is no limitation on God’s power but a statement of the internal logic of God. William of Ockham (1285-1347), the inventor of “Occam’s Razor,” made an important distinction between God’s absolute power and his power expressed in the order of the universe he creates.[6] God has the absolute power to suddenly change the relation between the sun and the Earth, or to turn a senator into a duck. But God’s “ordered power” prevents him from doing either. God’s absolute power is absolutely unlimited, but he limits his own ordered power. Ockham’s idea was, by the way, an essential step toward science as well as an essential basis of theology. People often pray for rain. It is within God’s absolute power to bring rain to their locality without changing the weather patterns of the whole Earth, but it is not within his ordered power. If God suddenly drenched one field or town with rain, that would have an effect on the weather throughout the world.[7]

How, then, do Christians evaluate miracles, how do they separate miracles from marvels, delusions, frauds, and mistakes? On the basis of the reasonableness of the testimony. Before the 1700s, or at least the 1600s, the truth of events was weighed according to testimony, a method still used in some fields such as the legal system. The reasonableness of testimony arose from the reliability of the witnesses. One of the greatest English historians of all time, Bede (673-735) poses a problem for historians. On the one hand, his history of the English people up to his own time has been confirmed in most details by modern history and archeology; on the other hand, it contains numerous accounts of miracles that Bede carefully examines and some of which he declares to be valid. How, modern historians fret, can Bede be so reliable in one way and so unreliable in another? Contemporary historians usually evade the question by patronizing Bede, saying that he was influenced by his culture, as if contemporary historians aren’t influenced by theirs. Bede evaluated reported miracles on the basis of whether they were reasonably reported by reliable witnesses.[8]

In the 1700s rationality shouldered reasonableness aside and replaced it with narrower “rationality.” Events were to be judged true or false, not on the basis of testimony, but on the basis of repeatable experimentation. Rationality works well in the natural sciences, but not well in history, for historians still judge events by the reliability of witnesses. Still, most contemporary historians now simply assume that a witness reporting a miracle (or even a wonder) is unreliable. Humans still observe miracles, but unlike other events, miracles are not allowed as evidence. Yet “it can’t happen” is true only if nothing can happen except things explained by physicalism, and physicalism isn’t a demonstration or even an argument, it’s the axiom (the unsupported assumption) that everything real is physical, followed by the corollary that nothing that isn’t physical is real.[9]  No miracles are allowed because no miracles are possible. This undemonstrated assertion does not explain miracles away at all.


[1] Thomas Crean, God Is No Delusion: A Refutation of Richard Dawkins (San Francisco: Ignatius, 2007), 53.
[2] Crean, 52-53.
[3] Gregory E. Ganssle, “Dawkins’ Best Argument against God’s Existence,” in Paul Copan and William Lane Craig, eds., Contending with Christianity’s Critics: Answering New Atheists and Other Objectors (Nashville, TN: B&H, 2009), 74-86.
[4] AARP: The Magazine (January/February 2009): 50-64.
[5] Ibid. 52.
[6] Potestas absoluta and potestas ordinata.
[7] Christoph Cardinal Schönborn, Chance or Purpose? Creation, Evolution, and a Rational Faith (San Francisco: Ignatius, 2007), 78-83, on contingency and continuous creation. Since new things appear every moment, creation is obviously developing, so God must be developing it. But this does not mean that God is himself developing, as Process Theology (which began with Alfred North Whitehead in 1928-1929) states. Everything past, present, and future is contingent on God, who is continuously creating. One might say he foresees every development in nature—or better, he acts with every development in nature.
[8] Rick Kennedy, History of Reasonableness (Rochester, NY: University of Rochester Press, 2004).
[9] A mathematician suggests that “ideas from information theory need to be taken more seriously by the physicists and cosmologists. The physical universe and the objects in it could be a manifestation of some other reality in which physical objects are structured the way software is…..None of the mathematics used by physicists has its truth predicated on its reducibility to statements about matter and energy.”  Communication to the author by Professor  S. Gill Williamson.


Jeffrey Burton Russell is Professor Emeritus of History at the University of California, Santa Barbara, and author of Exposing Myths About Christianity: A Guide to Answering 145 Viral Lies and Legends.

View the archived discussion of this post

This article is now closed for new comments. The archived comments are shown below.

Page 1 of 1   1
Lou Jost - #81413

June 28th 2013

“A miracle is a supernatural event, which in its very nature is inaccessible to science, and any scientist honestly open to inquiry into natural phenomena—which is her vocation—will admit as much.” Not true. Even a one-time event leaves tracces that are potentially verifiable. And exceptions to “laws” are carefully studied when possible, because they are important pointers toward better “laws”.

This excerpt misses the central reason why most scientists don’t believe in miracles: the evidence for them is not good enough to outweigh their extreme improbability. This is the same reason why many Christians don’t believe in the miracle stories of other religions, or of contemporary psychics or gurus.

Eddie - #81418

June 29th 2013

Hi, Lou.

Yes, a miracle might leave traces; and therefore the fact that the unusual event occurred would not necessarily be inaccessible to science.   But I suspect that the author’s meaning is that the cause of the event will be inaccessible to science.  Science can provide explanation only for events that fall under causal regularities.  And the Resurrection, and the creation of the world out of nothing, by their very nature do not fall under causal regularities.  There are no “general principles governing Resurrections” and “general principles governing emergence from nothingness” (I mean real nothingness, not the so-called “quantum vacuum” which is not “nothing”).  

So there can be no scientific explanation of “... and there was light” (Gen. 1.3).  And even if we can fantasize a set of “laws” (governing reversals of cell-death or the like) that might produce resurrections under certain rare circumstances—say, once every 10 billion deaths or so—the context of the Biblical story makes it pretty clear that the Resurrection is to be understood as a special divine action focused on the unique individual Jesus, not as something that could happen to any random person due to natural causes.

When you say “most scientists don’t believe in miracles,” you seem to mean:  “Most scientists don’t believe that the events that are alleged to be miraculous have actually happened.”  I.e., most scientists don’t believe that Jesus rose from the dead, Moses parted the waters, Joseph Smith found golden tablets, etc.  But then the question is why they don’t believe this, given that these are represented as unique events, not following from any natural laws that scientists could hope to discover.

If they don’t believe it because they think the documents are unreliable, or the eyewitnesses were uncritical and easily duped, etc., then they are not making scientific judgments, but historical or literary judgments.  And that’s fine—but you can’t claim expertise in natural science as the basis of a historical or literary judgment. 

So if your scientists are saying:  ”If I thought these events were reliably reported, then I would grant that they happened, and then I would have to at least consider the possibility that they had a supernatural rather than a natural cause, but I think (speaking as a historian/literary critic) they aren’t reliably reported, so I don’t have to bother with the scientific investigation of their causes,” that is one thing; but if they are saying, “I don’t believe these events were reliably reported, because as a scientist I know that such events would require a breaking or suspension of known natural laws and therefore could not possibly have happened,” then they are offering an a priori commitment regarding what kinds of events can happen in the world—a metaphysical commitment—as an authoritative deliverance of “science.”  And that’s an illegitimate line of argument.


Lou Jost - #81424

June 29th 2013

Eddie, my own position (and I think this is the most common position among scientists) is somewhere between your two alternatives. We can’t rule out the possibility of miracles a priori, but we can say that if they happened as reported, fundamental laws that are apparently universal must have been violated. We would therefore hold reports of such an event to a higher standard of evidence than reports of an event which did not violate so many of our observations about the world. It is legitimate to bring our knowledge of the world to textual analysis. As I mentioned. everyone does this. If a kid says “A monster ate my homework”, the teacher should be skeptical, even though it is logically possible.

Now suppose a teacher tells us a story  that had been passed down at his school from earlier teachers. The story claimed that a teacher had a kid in his class who said “My homework paper turned into five thousand fish”. Well, that is even worse, because while monsters might exist in some unexplored ecosystem and might have gotten lost and wandered into suburbia USA, the fish story violates so many fundamental laws that we would need very strong evidence indeed, and the fact that this is a second-hand story adds to the fishiness of it. Even if the kid actually had come to school with five thousand fish, and swore it happened, and even if there were still pictures of the fish in the yearbook, we’d still be justifiably skeptical. And then when the story mentions that the kid’s entire family of twelve are commercial fishermen, we might safely toss the story, even if everyone involved in the story is perfectly honest and believes the story. Even the kid might have been honest—-maybe he left his homework somewhere and his family happened to dump their latest catch on top of the kid’s homework, and the kid, being superstitious, just assumed that his homework turned into five thousand fishes.

Eddie - #81446

June 29th 2013


I understand your position.  I don’t find it unreasonable.   And I agree that in practice we all hold claims of unheard-of events to a higher standard than others.

In fact, your position seems very close to the “official” position of David Hume, as outlined in his famous essay on miracles.  But I wonder if you don’t also, deep down, hold to Hume’s “unofficial” position, i.e., that in practice, the bar is always so high that no miracle claim can ever be allowed.

Let me explain.  Most editions of Hume’s famous essay leave out a very long footnote included by Hume.  The footnote discusses cases of reported modern miracles.  In the note, we see that Hume dismisses some miracle reports given, not just by uneducated country bumpkins or by men known to be liars, or to have wild imaginations, or to lack critical faculties, or to be easily taken in by frauds,  but by men of unquestioned personal integrity, of great sobriety and circumspection, and of great intelligence and critical ability.  If the reports of such men cannot be trusted, then whose reports can?  Hume in practice sets the bar for testimony so high that no can jump it.  So in practice we never have to acknowledge the occurrence of a miracle, or even to say that it is more likely than not that the miracle has occurred.  We can simply carry on as if miracles do not occur.  The purported neutrality and open-mindedness of Hume turns out to be a sham, and it would be preferable if he were as honest as Voltaire or Spinoza and simply discounted miracles altogether.

Now I’m not accusing you of being as doctrinaire as I think Hume secretly was, but merely pointing out that an apparently healthy skepticism can easily mask a hidden dogmatism.  I’m wondering if you would ever accept any written or oral report as strong evidence for a past miracle, or whether you would only accept miracles you had witnessed with your own eyes, discounting all secondhand reports as unconfirmable, tainted by the prior religious beliefs of the reporter, perhaps due to optical illusions or insufficient sleep or hallucinogens, etc.

Lou Jost - #81448

June 29th 2013

Eddie, thanks for the thoughtful comment. I do set the bar pretty high when it comes to “big” miracles”, like Islam’s miracle of the moon splitting into two halves. There would have to be independent testimony from various unrelated cultures. Even then, I would look very very hard for the possibility of optical illusion due to some weird atmospheric effect. Probably the only thing that would convince me it really happened would be evidence from an Apollo mission that could only be explained this way. And even then, I would prefer a natural explanation in terms of a giant asteroid impact or something like that. But I don’t think this is being dogmatic, it is just being reasonable.We know that such miracle reports are frequently exaggerated or have their origin in justifying certain beliefs rather than in accurately reporting events. And people are very, very gullible.

I’ll add that I spent many years doing experiments in parapsychology. Even got grants, had experimental facilities at two universities, and confronted hostile skeptic scientists in my departments. I was interested in what paranormal phenomena would teach us about the nature of time, and about determinism, if such phenomena had been real. As it turned out, the evidence just wasn’t good enough. But I gave it a good shot at great personal cost. I am one of the least dogmatic scientists I know. It makes me smile when people think I am dogmatic now….

As part of my study of paranormal phenomena, I learned magic and misdirection under a psychology prof who was also a magician. I could be a miracle-worker in many parts of the world if I wanted to!!!! It is really easy to fool people. Especially when they want to be fooled. After people have been primed, one can almost just sit back and let random natural stuff happen, and people will fall into miraculous interpretations at the slightest nudging. Oh, and people are TERRIBLE observers, and even worse at remembering their observations. Learning magic is a great way to reduce the weight you put on first-hand observation…and you can forget about second-hand observations….

Lou Jost - #81451

June 29th 2013

To  take the resurrection and ascension seriously, I would need at least a couple of non-anonymous eyewitness accounts. I’d want at least one of the eyewitness accounts to be by a Roman or other non-follower of Jesus. That is a minimum set of requirements that I think most people would ask for. I am not saying this would convince me, but it would get me thinking.

Eddie - #81501

July 1st 2013


I partly agree with you.  If someone were to say that he had “historical proof” of the Resurrection, and meant that as an argument that should compel all unbelievers to come into the Christian fold, then I would say that the unbeliever could reasonably demand confirmation from non-Christian witnesses; e.g., “Show me a Syriac or Greek or Latin document written by a Jew or pagan confirming the event.”

But I suspect that many Christians do not conceive of themselves as possessing a “proof” of the Resurrection that would be equally convincing to non-believers as believers.  I suspect that many of them  believe in the Resurrection because they believe that they have, in one way or another (visions, voices, awareness of a presence, etc.), experienced Jesus as alive.  However, they know that this sort of subjective certainty will not convince an unbeliever, who is looking for tangible evidence.  So they would not speak of “proof” in the forensic sense.

Unfortunately, a certain type of fundamentalist does not exercise such epistemological restraint, and so you get a whole literature of the “evidence that demands a verdict” type.  Most Christians are content to accept a combination of testimony (not regarded as “hard proof” but as trustworthy) and personal religious experience as enough assurance for things like the Resurrection.  The fundamentalists push the issue of proof too aggressively.  They thus end up almost endorsing a version of scientism, because they argue, in effect, that the kinds of positive proof demanded in “the scientific study of history” can be demanded of the Bible, and that the Bible should deliver.  And then the atheists lash back about the inadequacy of the proofs.  At that point the discussion becomes interminable, and pointless for me, as I’m not interested in any proofs or disproofs of the sort being offered by either side.  

Contemporary miracle claims are another matter.  At least in some cases, they are testable by scientific or at any rate observational means.  A few years back it was reported that a Hindu idol at a temple had been drinking the milk from its bowl.  The milk appeared to have been drained the next morning.  This supposedly happened many times over an extended period.   It would then have been possible to set up hidden cameras, etc. to watch what happened.  Maybe the god was drinking the milk, or maybe the whole thing was a pious fraud by a temple functionary.  The camera would tell the story.  Then one could speak of proof or disproof.

But miracles reported in ancient sources—there is no way of obtaining proof by modern standards.  So religious people are foolish to claim they can provide it.  That does not make it irrational for them to believe what they believe; it only makes it irrational for them to go around trying to bully people with “proofs.”

Lou Jost - #81518

July 1st 2013

Eddie, thanks for substantially agreeing. I think even fairly progressive Christian scholars often overstate the strength of the historical evidence for things like the resurrection. Both P and Wright are examples. I am pretty sure they believe for other (perhaps very subjective) reasons, as you said, and then see the historical documents through this lens of subjective certainty. These people “know” their particular brand of god exists (as Jon asserts on this thread), and so they are not very skeptical of his associated wonders. This happens in all religions and cults. The milk-drinking Ganesh statues you mention are good non-Christian examples.

Eddie - #81523

July 1st 2013


Agreed, but I again point out that atheists, materialist, reductionists etc. also view historical documents, scientific data, etc. through lenses of certainty of their own, and can be just as metaphysically or epistemologically biased.  I think the hostility to teleology in nature is an example; I think the hostility goes beyond what actual discoveries can account for and has deep subjective roots; but it is best to discuss that on the other thread where we have already started in on teleology.


GJDS - #81469

June 30th 2013

Discussions on miracles can become so circular that they become pointless; i.e. by definition God has the power to perform miracles, but we do not have such power, to initiate them nor ‘hold them’ in place for testing. People believe they have experienced miracles, and the final judgment can only rest with them. It is also true that cultures have recorded magic and various such ‘arts’ and people have also believed in them. Those instances that could be examined have in most cases shown to be bogus.

It is also true that where possible, the Church has gone to great lengths, over long periods, to obtain as much data and information as possible, before accepting any miraculous event. This is an important point for these discussions. It is wrong to believe that Christians regard miracles as some sort of ‘event’ for public attention. Instead we all regard such events as rare and an indication of the workings of faith.

We need to be reminded that the distinction between magic and miracles (miracles done for people’s well being and for faith, as opposed to magic done to convince people of the magician’s power and importance) is discussed in Acts 8:9-24. We can see here that a person who was skilled at magic felt that he could do more of this if he could somehow become as the Apostles – but he could not understand the purpose of the healings performed by the Apostles – he understood instead how he could exert influence and power over people. It is worth contemplating this account, when this topic comes up for more disputation.

Jon Garvey - #81473

June 30th 2013

I can’t find any data online that actually surveys scientists to find whether a majority do reject miracles, except for India, which is a restricted subject-base as much as all-US studies are. Any claims about what a majority believe are merely anecdotal without sociological data.

Even interpreting the surveys that claim most scientists don’t believe in God is fraught with hazard. A nice recent article questioning the commonly received wisdom here. The poor construction of surveys, in one way or another, is notable and demonstrable - if one looks at the research rather than the spun results: equally true in all walks of life.

But by all accounts there seems to be at least a sizeable minority of (elite) scientists who believe in God, or even a majority if you count “some spiritual beliefs”. Of those a good proportion no doubt do believe miracles are possible. One then has to ask whether they are worse scientists because of that by definition, or better, perhaps because they’re in a minority who think more about metaphysics and philosophy instead of being blinkered by materialistic assumptions. After all, do you expect to find broader intellectual education amongst elite scientists, or amongst somewhat less elite scientists?

One 2004 survey found that 74% of US physicians believe in miracles. Unfortunately, although the research surveyed religious background, it is not clear whether specifically religious physicians were questioned, or whether, say, atheists from a Catholic background counted as “Catholic”. Nevertheless, as a physician I can confirm that daily experience of real people and situations had a tendency to soften the hard edges of my colleagues’ materialism. Quite frankly, you’re a lot more likely to encounter a possible miracle in a doctor’s office than in a genetics lab, even before you start theorising about it.

Lou Jost - #81474

June 30th 2013

We’vebeen through the survey results in other threads. The vast majority of scientists in the UK, for example, do not believe in god. This is especially true of physicists and biologists. When “eliteness” has been measured, it is always found that the more elite scientists are less likely to believe in god. Yes, the surveys need to be read with caution, but the overall trend is clear.

Medicine is a messy subject with little control. It is hard to even define “miracle” in a healing context.


Page 1 of 1   1